Reinforcement learning for dialog management using least-squares Policy iteration and fast feature selection.
Lihong LiJason D. WilliamsSuhrid BalakrishnanPublished in: INTERSPEECH (2009)
Keyphrases
- reinforcement learning
- feature selection
- reinforcement learning algorithms
- temporal difference
- model free
- reinforcement learning methods
- policy iteration
- machine learning
- function approximation
- management system
- markov decision processes
- state space
- state action
- feature selection algorithms
- action selection
- support vector machine
- text categorization
- text classification
- model selection
- supervised learning
- feature extraction
- continuous state spaces
- feature set
- knowledge management
- knn
- classification accuracy
- user interface
- support vector
- mutual information
- conversational agents
- transfer learning
- optimal policy
- data management
- learning process
- feature space
- information systems
- learning algorithm